To make things a little more interesting, let’s turn this into a competition. Remember those scenarios?

  1. Softest landing - land on the moon with zero speed.
  2. Accurate landing - touchdown on the landing pad as accurately as possible.
  3. Level landing - land as level as possible.
  4. Minimal fuel usage - conserve as much fuel as possible.

Your goal is to optimize a lunar lander agent to land on the moon in the scenarios listed above. You will have a chance to train multiple RL models to perfect each scenario. The results will be sent to a leaderboard so you can see where you rank among other teams. You will also be able to see live video replay of agent training being done by other teams.

The website URLs to view the leaderboard and live video replay will be displayed on the Event Engine dashboard, similar to below. Check your dashboard to copy the appropriate links.

When you are ready to get started, continue on.